Spatial effects on hybrid electric vehicle adoption

• Develop spatial regression models to estimate spatial effects on hybrid electric vehicle adoption.
• Apply models to a case study based on the state of Ohio.
• Demonstrate direct spillover effects and spatial correlations of errors in vehicle adoption patterns.
• Demonstrate spatial correlations in temporal vehicle adoption as well.
• Discuss use of spatial models for simulating and predicting pockets of alternative vehicle adoption.

This paper examines spatial effects on hybrid-electric vehicle (HEV) adoption. This is in contrast to most existing analyses, which concentrate on analyzing socioeconomic factors and demographics. This paper uses a general spatial model to estimate the strength of ‘neighbor effects’ on HEV adoption—namely that each consumer’s HEV-adoption decision can be influenced by the HEV-adoption decisions of geographic neighbors. We use detailed census tract-level demographic data from the 2010 United States Census and the 2012 American Community Survey and vehicle registration data collected by the Ohio Bureau of Motor Vehicles. We find that HEV adoption exhibits significant spatial effects. We further conduct a time-series analysis and show that historical HEV adoption has a spatial effect on future adoption. These results suggest that HEVs may appear in more dense clusters than models that do not consider spatial effects predict.